onwards 0.19.1

A flexible LLM proxy library
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
//! Chat Completions API schemas
//!
//! These schemas match the OpenAI Chat Completions API specification.
//! See: https://platform.openai.com/docs/api-reference/chat

use serde::{Deserialize, Serialize};

/// Request body for POST /v1/chat/completions
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChatCompletionRequest {
    /// The model to use for completion
    pub model: String,

    /// The messages to generate a completion for
    pub messages: Vec<ChatMessage>,

    /// Sampling temperature (0-2)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub temperature: Option<f32>,

    /// Nucleus sampling parameter
    #[serde(skip_serializing_if = "Option::is_none")]
    pub top_p: Option<f32>,

    /// Number of completions to generate
    #[serde(skip_serializing_if = "Option::is_none")]
    pub n: Option<u32>,

    /// Whether to stream the response
    #[serde(skip_serializing_if = "Option::is_none")]
    pub stream: Option<bool>,

    /// Stream options (e.g., include_usage)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub stream_options: Option<StreamOptions>,

    /// Stop sequences
    #[serde(skip_serializing_if = "Option::is_none")]
    pub stop: Option<StopSequence>,

    /// Maximum tokens to generate
    #[serde(skip_serializing_if = "Option::is_none")]
    pub max_tokens: Option<u32>,

    /// Maximum completion tokens (newer parameter)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub max_completion_tokens: Option<u32>,

    /// Presence penalty (-2.0 to 2.0)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub presence_penalty: Option<f32>,

    /// Frequency penalty (-2.0 to 2.0)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub frequency_penalty: Option<f32>,

    /// Logit bias for tokens
    #[serde(skip_serializing_if = "Option::is_none")]
    pub logit_bias: Option<serde_json::Value>,

    /// Whether to return log probabilities
    #[serde(skip_serializing_if = "Option::is_none")]
    pub logprobs: Option<bool>,

    /// Number of most likely tokens to return
    #[serde(skip_serializing_if = "Option::is_none")]
    pub top_logprobs: Option<u32>,

    /// User identifier for abuse tracking
    #[serde(skip_serializing_if = "Option::is_none")]
    pub user: Option<String>,

    /// Random seed for deterministic sampling
    #[serde(skip_serializing_if = "Option::is_none")]
    pub seed: Option<i64>,

    /// Tools available to the model
    #[serde(skip_serializing_if = "Option::is_none")]
    pub tools: Option<Vec<Tool>>,

    /// How to choose which tool to use
    #[serde(skip_serializing_if = "Option::is_none")]
    pub tool_choice: Option<ToolChoice>,

    /// Parallel tool calls setting
    #[serde(skip_serializing_if = "Option::is_none")]
    pub parallel_tool_calls: Option<bool>,

    /// Response format specification
    #[serde(skip_serializing_if = "Option::is_none")]
    pub response_format: Option<ResponseFormat>,

    /// Service tier (auto, default, etc.)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub service_tier: Option<String>,

    /// Additional fields not explicitly modeled
    #[serde(flatten)]
    pub extra: Option<serde_json::Value>,
}

/// A message in the conversation
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChatMessage {
    /// The role of the message author
    pub role: String,

    /// The message content (can be string or array of content parts)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub content: Option<MessageContent>,

    /// Name of the author (for function/tool messages)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub name: Option<String>,

    /// Tool calls made by the assistant
    #[serde(skip_serializing_if = "Option::is_none")]
    pub tool_calls: Option<Vec<ToolCall>>,

    /// Tool call ID (for tool messages)
    #[serde(skip_serializing_if = "Option::is_none")]
    pub tool_call_id: Option<String>,

    /// Additional fields
    #[serde(flatten)]
    pub extra: Option<serde_json::Value>,
}

/// Message content - either a string or array of content parts
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(untagged)]
pub enum MessageContent {
    Text(String),
    Parts(Vec<ContentPart>),
}

/// A content part in a multimodal message
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(tag = "type")]
pub enum ContentPart {
    #[serde(rename = "text")]
    Text { text: String },
    #[serde(rename = "image_url")]
    ImageUrl { image_url: ImageUrl },
}

/// Image URL specification
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ImageUrl {
    pub url: String,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub detail: Option<String>,
}

/// Stream options
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct StreamOptions {
    #[serde(skip_serializing_if = "Option::is_none")]
    pub include_usage: Option<bool>,
}

/// Stop sequence - single string or array
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(untagged)]
pub enum StopSequence {
    Single(String),
    Multiple(Vec<String>),
}

/// Tool definition
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Tool {
    #[serde(rename = "type")]
    pub tool_type: String,
    pub function: FunctionDefinition,
}

/// Function definition for a tool
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct FunctionDefinition {
    pub name: String,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub description: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub parameters: Option<serde_json::Value>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub strict: Option<bool>,
}

/// Tool choice specification
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(untagged)]
pub enum ToolChoice {
    Mode(String), // "none", "auto", "required"
    Specific {
        #[serde(rename = "type")]
        tool_type: String,
        function: ToolChoiceFunction,
    },
}

/// Specific function choice
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ToolChoiceFunction {
    pub name: String,
}

/// Tool call in an assistant message
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ToolCall {
    pub id: String,
    #[serde(rename = "type")]
    pub call_type: String,
    pub function: FunctionCall,
}

/// Function call details
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct FunctionCall {
    pub name: String,
    pub arguments: String,
}

/// Response format specification
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ResponseFormat {
    #[serde(rename = "type")]
    pub format_type: String,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub json_schema: Option<serde_json::Value>,
}

/// Response from POST /v1/chat/completions (non-streaming)
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChatCompletionResponse {
    pub id: String,
    pub object: String,
    pub created: u64,
    pub model: String,
    pub choices: Vec<Choice>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub usage: Option<Usage>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub system_fingerprint: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub service_tier: Option<String>,
}

/// A completion choice
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Choice {
    pub index: u32,
    pub message: ChatMessage,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub finish_reason: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub logprobs: Option<serde_json::Value>,
}

/// Token usage information
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Usage {
    pub prompt_tokens: u32,
    pub completion_tokens: u32,
    pub total_tokens: u32,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub prompt_tokens_details: Option<serde_json::Value>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub completion_tokens_details: Option<serde_json::Value>,
}

/// Streaming chunk response
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChatCompletionChunk {
    pub id: String,
    pub object: String,
    pub created: u64,
    pub model: String,
    pub choices: Vec<ChunkChoice>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub usage: Option<Usage>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub system_fingerprint: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub service_tier: Option<String>,
}

/// A streaming choice
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChunkChoice {
    pub index: u32,
    pub delta: ChunkDelta,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub finish_reason: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub logprobs: Option<serde_json::Value>,
}

/// Delta content in a streaming chunk
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChunkDelta {
    #[serde(skip_serializing_if = "Option::is_none")]
    pub role: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub content: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub tool_calls: Option<Vec<ChunkToolCall>>,
}

/// Tool call in a streaming chunk
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChunkToolCall {
    pub index: u32,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub id: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    #[serde(rename = "type")]
    pub call_type: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub function: Option<ChunkFunctionCall>,
}

/// Function call in a streaming chunk
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ChunkFunctionCall {
    #[serde(skip_serializing_if = "Option::is_none")]
    pub name: Option<String>,
    #[serde(skip_serializing_if = "Option::is_none")]
    pub arguments: Option<String>,
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_deserialize_simple_request() {
        let json = r#"{
            "model": "gpt-4",
            "messages": [
                {"role": "user", "content": "Hello"}
            ]
        }"#;

        let request: ChatCompletionRequest = serde_json::from_str(json).unwrap();
        assert_eq!(request.model, "gpt-4");
        assert_eq!(request.messages.len(), 1);
    }

    #[test]
    fn test_deserialize_multimodal_content() {
        let json = r#"{
            "model": "gpt-4-vision",
            "messages": [
                {
                    "role": "user",
                    "content": [
                        {"type": "text", "text": "What's in this image?"},
                        {"type": "image_url", "image_url": {"url": "https://example.com/image.jpg"}}
                    ]
                }
            ]
        }"#;

        let request: ChatCompletionRequest = serde_json::from_str(json).unwrap();
        assert_eq!(request.messages.len(), 1);
        if let Some(MessageContent::Parts(parts)) = &request.messages[0].content {
            assert_eq!(parts.len(), 2);
        } else {
            panic!("Expected parts content");
        }
    }

    #[test]
    fn test_deserialize_with_tools() {
        let json = r#"{
            "model": "gpt-4",
            "messages": [{"role": "user", "content": "What's the weather?"}],
            "tools": [
                {
                    "type": "function",
                    "function": {
                        "name": "get_weather",
                        "description": "Get the weather",
                        "parameters": {"type": "object", "properties": {}}
                    }
                }
            ]
        }"#;

        let request: ChatCompletionRequest = serde_json::from_str(json).unwrap();
        assert!(request.tools.is_some());
        assert_eq!(request.tools.unwrap().len(), 1);
    }

    #[test]
    fn test_serialize_response() {
        let response = ChatCompletionResponse {
            id: "chatcmpl-123".to_string(),
            object: "chat.completion".to_string(),
            created: 1234567890,
            model: "gpt-4".to_string(),
            choices: vec![Choice {
                index: 0,
                message: ChatMessage {
                    role: "assistant".to_string(),
                    content: Some(MessageContent::Text("Hello!".to_string())),
                    name: None,
                    tool_calls: None,
                    tool_call_id: None,
                    extra: None,
                },
                finish_reason: Some("stop".to_string()),
                logprobs: None,
            }],
            usage: Some(Usage {
                prompt_tokens: 10,
                completion_tokens: 5,
                total_tokens: 15,
                prompt_tokens_details: None,
                completion_tokens_details: None,
            }),
            system_fingerprint: None,
            service_tier: None,
        };

        let json = serde_json::to_string(&response).unwrap();
        assert!(json.contains("chatcmpl-123"));
    }
}